Source: UNIVERSITY OF CALIFORNIA, RIVERSIDE submitted to NRP
NEW TREE SPECIES MAPS IN SOUTHERN CALIFORNIA USING GOOGLE EARTH
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1006645
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Jun 23, 2015
Project End Date
Sep 30, 2019
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
UNIVERSITY OF CALIFORNIA, RIVERSIDE
(N/A)
RIVERSIDE,CA 92521
Performing Department
Earth Sciences
Non Technical Summary
The primary objective of this proposal is to produce high resolution maps of tree species in coastal southern California south of latitude 35º N, using the most recent imagery on Google Earth. This new baseline will provide a foundation for studies of spatially explicit historical vegetation change, as well as assessing future change. Existing information on species distribution has proven inadequate to manage exotic pests and pathogens, changes in wildfire patterns, and large-scale mortality events. A new forest spatial database would contribute to both the management of these problems and the basic science required understand how southern California forests may change in the future. Maps from this project will be archived at the Center of Conservation Biology (CCB). A secondary objective will be to illustrate the spatial-temporal approach with case studies of forest change.
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
12306131070100%
Goals / Objectives
In the southern California region it is vital to develop baseline data to evaluate the dynamics of ecosystems over long time and broad spatial scales. The Los Angeles-San Diego megalopolis of ca. 25 million people is surrounded by undeveloped wildlands in public landholdings under jurisdiction of the US Forest Service, National Park Service, and California State Parks, as well as in municipal and county jurisdictions. Intense public pressure on natural lands, including recreation, development of cities within National Forest Boundaries, and fire suppression have historically led to unintended impacts on ecosystems. At the institutional level, conservation legislation mandates spatial databases of plant species. There is also interest in vegetation change from new or yet to come pathogens and invasive species, presently seen in incipient damage and mortality to red oaks by the newly introduced Goldspotted oak borer (Coleman and Seybold 2008).Scientific hypotheses can also be tested at regional scales. In ecological science, the recent interest in global warming has shifted paradigms from the understanding steady-state functioning and dynamics of ecosystems to a new direction on future change of ecosystems. But one is dependent on the other. The question is whether it is prudent to bypass empirical research toward modeled vegetation experiments that are also based on models of climate change. Where does empiricism begin and end? These predictive efforts--time-to-space substitution--risk similar hazards of time-to-space substitution which plague retrospective studies similarly challenged by historical data. To diminish the limitations of both approaches it is important to consider historical documentary evidence for rigorous calibration for models predicting vegetation change. The hindsight of vegetation history, while partially understood, provides insight and empirical constraint to the foresight of vegetation change in modeling experiments. It is vital to renew efforts to gather historical information from the original primary sources such as written documents and time-series aerial photographs to assess or test broad scale long-term change in species distribution or calibrating future vegetation change models, following the example of long-established protocols for validation of Global Climate Models against proxy climate records at geologic time scales (e.g., CLIMAP 1976; COHMAP 1988).Vegetation surveys have been undertaken in southern California ever since the area was first colonized by Europeans in the late 18th century (Minnich 2008). Field descriptions were recorded in the Spanish explorations of 1769-1776, the 1848-55 US Mexican Boundary Survey, 1852-57 Pacific Railroad Survey, and 1861-64 State Survey of botany. The Spanish journals of the Portola and Anza expeditions have proven to be a valuable baseline to document the conversion of California flower fields to invasive exotic annual grassland (Minnich 2008). Spanish texts also record 25 forest species at locations between San Diego and San Francisco that are consistent with modern species ranges (Minnich and Goforth in review). The persistence of forest and woodland distributions across California since Spanish explorations illustrates that native assemblages have responded slowly to climate change. Spanish documents were written in the Little Ice Age (1500-1850) after which global temperatures have increased >1.0º C. Historical accounts of forest in the San Bernardino Mountains and in similar ecosystems of northern Baja California Mexico are also consistent with modern distributions (Minnich 1988; Minnich and Franco Vizcaino 1998). However, these surveys document cases of consistency with modern species ranges, not spatially explicit change nor transformations in vegetation properties. Botanical collections have been taken throughout California since the mid-19th century and presently archived on-line in the Consortium of California Herbaria (CCH, http://ucjeps. berkeley.edu/consortium/), Calflora (http://www.calflora.org/) and the VTM Survey http://vtm.berkeley.edu/; Griffin and Critchfield 1972). While botanical collections are taxonomically indisputable, the database is too sparse for inventory of vegetation change. Botanical samples are site-specific, and collections represent novel or unusual distributions, often outposts beyond of regional distribution, and often biased by access along roads and trails. Early botanical collections are poorly located and later collections were based on township and range system, with location error of 100 to 1000 m. Collections have historical value, especially in the spread of introduced invasive species (Minnich 2008).Polygon records of vegetation maps bring insight into historical vegetation change not possible from other data sources such as journals of historic expeditions and botanical collections. Depending on boundary criteria, maps are comprehensive in the rendition of spatially explicit species ranges. Most importantly, the production of maps based on aerial photographs and space imagery can be replicated in future studies using these original records. Map criteria can also be modified to estimate other properties such as regional cover and abundance of species which is vital data in vegetation and species inventory, forest and wilderness management, fire hazard analysis, land use planning, as well as long-term vegetation change. Map coverages can be correlated with terrain properties (slope, aspect, elevation) rendered in digital elevation models (Minnich 2001), as well as climate, bedrock, lithology and other parameters.The first vegetation maps of southern California were published in 1887 under the California State Board of Forestry (Kinney 1887). This effort was soon followed by maps of the San Gabriel, San Bernardino and San Jacinto Mountain Forest Reserve surveys by the US Geological Survey (Leiberg 1899, 1900) and the eastern San Bernardino Mountains (Grinnell 1908). The first formal survey of California vegetation statewide was undertaken by the Vegetation Type Map (VTM) Survey from 1929-34 which published detailed maps by topographic quadrangle (Weislander 1938; Colwell 1977; http://vtm.berkeley.edu/). While these map inventories provide an invaluable historical data, the documentation of long-term change requires replication of VTM methods, rather than original vegetation data.Vegetation change can be rigorously documented using original data from aerial photographs dating to the early 20th century, and space imagery since the 1970s that can be classified by any methodological protocol. Specifically, the 80-year time-span of archived imagery approaches the average lifespan of many tree species. Sequential georeferenced imagery provides original records of vegetation at specific moments of time, and now spans nearly a century since the onset of aerial overflights in the late 1920s and 1930s. Aerial photograph flights have been undertaken over most of coastal southern California every decade and are available at government institutions, and on-line in Earth Explorer USGS (http://earthexplorer.usgs.gov/). The creation of vegetation maps from modern databases provides added value to historical photographs, as well as a baseline for future vegetation change. This can be accomplished efficiently by use of Google Earth on-line software.
Project Methods
Vegetation analysis using Google Earth.High resolution vegetation maps have been made possible by the recent technical revolution in Google EarthTM (http://www.google.com/earth/), an on-line software program that represents a union of remote sensing and geographic information systems (GIS). [J1]The achievement of Google Earth is the ability to electronically map (digitize) geographic data directly onto imagery. This eliminates several steps (and costs) in the production of geographic data: (1) original analysis of imagery independently of base maps; (2) transfer of data from imagery to base maps; (3) digitizing base map data into a geographic information system; and (4) reduction of data loss in these transfer processes. Imagery properties and resolution on Google Earth are ever-improving, permitting the identification of tree species on the basis of resolution, morphological properties, color, and shadows. The historical aerial photograph georeferencing feature of Google Earth allows for the study of the fate of forest ecosystems back to ca. 1994. Earlier historical photographs in Earth Explorer can be manually scaled to Google Earth for data transfer. This imagery and mapping system also speeds the transfer of information to other scientists and managers, allowing for rapid adoption of tree distribution maps by a wide range of resource management professionals.The philosophy of this project is that vegetation mapping is an intellectual process of manual interpretation of imagery rather than through computer scanning based on plant spectral properties. While computer scanning is efficient, the efficacy of vegetation databases is compromised by taxonomically ambiguous spectra. All plants comprise chlorophyll foliar surfaces with broadly overlapping reflectance spectra. Signal is made further amorphous by complex geometry of tree species canopies, and the fundamental observation that plant assemblages change continuously in species composition and morphology across landscapes. Even with the putative objectivity of computers, the correction for computer error is not. It is better to have intellectual synthesis with error, than the correction of computer error, based on unknown criteria. Human error can be corrected in future studies. In the Google Earth analysis, species will be mapped one at time, rather than by regional mapping of several species. This protocol allows for the focus on morphological properties that permit diagnostic identification of each species. In addition, the region is observed repeatedly, which encourages the recognition of "surprises," and omissions. Identifications are deduced based on the finite number of species in each of the life form classes. The "time-series aerial photo feature" was used to establish the best imagery for each species. Google Earth imagery will be scanned comprehensively in an "X-Y" format across the landscape. Imagery is examined at the smallest possible scale within the resolution of trees (c. 1:3500) to efficiently scan maximum area; when a population is suspected or encountered, Google Earth can be scaled to higher resolution (down to 1:150) to confirm identification. Near-ground inspection is not possible due to pixelation of imagery. Although Google Earth imagery is monoscopic, the tilt feature in Google Earth was used for three dimensional viewing of terrain and rock substrate properties of individual populations. Terrain is also visualized while imagery is in motion. Resolution does not permit the use of the tilt function for 3-D viewing of trees. Imagery is examined looking "south," (south to top) against sunlight angle. This allows the observation of tree morphology independently of shadows, to avoid repetitive illusion between crowns and shadows. A "north view" (north to top) is taken to examine tree shadow morphology. Tree ranges will be defined at a 10% cover threshold consistent with previous work (Minnich 1987; Minnich and Everett 2001; Minnich and Goforth 2011) and California resource policy definitions (see Anon. 2010). Using Google Earth over the past 5 years, I have published mapping surveys including palm oases and tree distributions in southern California and Baja California, and chaparral sky islands along the Baja California Peninsula (Minnich et. al 2011a, 2011b, in press). Figure 1 shows the distribution of Quercus agrifolia in northern Baja California to illustrate the detail and resolution of the Google Earth imagery and corresponding map precision. Presently, I am presently writing a book entitled Fire selection: the consequences of fire suppression along the southern California--northern Baja California borderland that uses Google Earth databases on conifer forest distributions prepared for the eastern half of coastal southern California, including the eastern San Gabriel Mountains, San Bernardino Mountains, San Jacinto Mountains and mountains of San Diego County. In this project I plan to expand this conifer tree species database westward to the mountains of Ventura and Santa Barbara Counties and Point Conception, as well as the southern California islands (cf. Minnich 1980). I have already published maps of southern California coniferous tree distributions (Minnich and Everett 2001) but Google Earth digitizing software will permit the development of superior maps. New maps will also include hardwood tree species. These maps represent an extension of published maps for conspecific tree species in Baja California. To back up files, vegetation maps produced under this proposal will be archived in the GIS lab at the Center for Conservation Biology. These maps are also backed up by original imagery used in this analysis, the raw data also permitting alternative mapping protocols depending on the nature of research.[J1]Citation

Progress 06/23/15 to 09/30/19

Outputs
Target Audience:The target audience is academic, amateur and the public at large. Global warming has become the dominant scientific paradigm in vegetation science, shifting the focus from the understanding of steady-state ecological functioning and dynamics to future change of ecosystems. But one is dependent on the other. An important question is whether it is prudent to bypass empirical research toward futuristic modeling experiments that are also based on another layer of climate change modeling. Where does empiricism begin and end? These predictive efforts--time-to-space substitution--risk similar hazards of time-to-space substitution which plague retrospective studies similarly challenged by historical data. The specific hypothesis at hand is whether future climate change will drive plant species and assemblages both poleward and to higher elevations. Without real time verification, vegetation forecast models are an untested exercise; proper science requires empirical testing of the global warming hypothesis. Mapping the complexity of nature at high resolution has recently been made possible using Google Earth which permits rapid efficient inventory both spatially and temporally. Google Earth has also become the library of aerial imagery for the world, and this database will grow rapidly in the time frame of 21st century global climate model simulations. The maps in this atlas should be viewed as a single time slice in an evolutionary process of building upon previous work, through continual refinement with improvement of imagery quality and for long-term vegetation change. Vegetation science should follow the map-making tradition of bedrock substrate inventory in the geological sciences since William Smith in the early 19th century. The maps in this research can be refined or redefined for a wide range of research hypotheses because this analysis is shown with the original vegetation on Google Earth imagery. Revisions can be efficiently undertaken because aerial imagery record is archived as a library on the Google Earth platform, and new data can be entered electronically. The maps are the first comprehensive baseline since the Vegetation Type Map Survey (VTM) of 1929-34, and the first baseline for forest and woodland vegetation on the Baja California peninsula. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The methods and approaches of this research can be applied anywhere because of Google Earth's comprehensive global coverage of imagery. The analysis in this atlas led to an evolutionary development of methods that would permit broad scale inventory of tree species anywhere. While this effort did not directly lead to training and professional development at UCR, there is great potential to expand professional training elsewhere. Every region in the world has ecologists with skilled local background in imagery interpretation. How have the results been disseminated to communities of interest?It is the intent to disseminate the results of this research in a single electronic book to test and refine maps using a "community on-line atlas" using the Google earth platform. In an electronic format that georeferences this vegetation polygon analysis with the original vegetation data on Google Earth, maps can be taken into the field or in the laboratory using mobile electronic devices and computers. Maps can be reconstructed and redefined at high resolution for individual purposes, and with the history function one can evaluate the fate of individuals and plant assemblages over time. Maps can be examined in the field using a mobile device or in the laboratory. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? The atlas is a compilation of unprecedented high resolution maps of 85 tree species in the three Californias along the North American Pacific coast--California south of San Francisco, Baja California and Baja California Sur--interpreted from Google EarthTM imagery. These species represent diverse plant assemblages found in the western North America including California oak woodland, Mexican oak woodland, riparian forest, palm oases, closed-cone conifer forest, mixed evergreen forest, mixed conifer forest, subalpine forest and pinyon-juniper woodland. Google Earth is transformative platform for broad scale ecological research. The capacity of virtual scaling permits the viewing of vegetation at high resolution both at large scales for broad spatial pattern, and local scales for taxonomy, population characteristics, and map accuracy. Imagery quality permits the identification of tree species on the basis of resolution, morphological properties, color, and shadows. With virtual scaling, maps can be created and examined for relationships between broad scale distributions and local populations. This atlas contributes a new kind of map. This survey based on an analysis connects tree distributions with of aerial imagery archived on the Google Earth platform. Simultaneously, one can both see boundary data of species as an overlay of imagery that shows the distribution and visual properties of that species. The objectives accomplished include: (1) the extension of California tree distributions to their limits in Mexico; the documentation of a broad transition of California species in Mediterranean climate to Mexican species in summer rain climates of Baja California; the realization of a "living atlas" on an ideal platform of Google Earth, for correction, revision, and to trace future vegetation change from climate change, disturbance, pathogenic perturbations; the capacity of compare species dynamics in two countries with divergent land policies from which data can be used to modify the use of the biota in both countries; and to present a methodological approach with Google Earth that can be applied globally. For each species, new maps with unprecedented resolution are presented with discussion of the late Cretaceous and Cenozoic fossil record, patterns of disturbance, especially from fires and floods, and changes in distribution since the last ice age maximum 20,000 years ago. The final maps are generally consistent with the VTM survey 80 years ago. However, this analysis also resulted in 31 range expansions of well-known tree species that are unknown to science, largely because Google Earth imagery provides high resolution data in inaccessible terrain. In ecological science, recent interest in global warming has shifted paradigms from the understanding steady-state dynamics of ecosystems toward future change of ecosystems. But one is dependent on the other. An important question is whether it is prudent to bypass empirical research toward modeled vegetation experiments that are also based on models of climate change. Predictive modelling efforts--space-to-time substitution--risk similar hazards of time-to-space substitution which plague retrospective studies similarly challenged by historical data. To diminish the limitations of both approaches it is important to consider historical evidence for calibration of models predicting vegetation change. The Google Earth platform allows for spatially-explicit, time-series inventory of these research questions at regional scales. Natural ecosystems can be appraised probabilistically, accepting that nature does not follow rules without uncertainty by exhibiting some degree of randomness in space and time. The testing of long-term vegetation change would immeasurably benefit by the posting of the earliest aerial photographic record on Google Earth. Since the earliest imagery of the Californias dates to the 1920s and 1930s such an effort would extend the California baseline nearly 60 years beyond present postings. For long-lived forest species, an enlarged time line offers legitimate testing on the effects of global warming on distributions including the separation of steady states from directional change of permanence. By 2100 the Google Earth imagery library would comprise a 170 year baseline. The complexity of future forests will be ineffectually modeled by simulation methodologies, particularly when there is no empirical verification until the verification date decades into the future. The integration of Google Earth, field science and modeling experiments would the best of all possibilities in the development of broad-scale ecological science that addresses the natural complexity of ecosystems and their future change.

Publications


    Progress 10/01/17 to 09/30/18

    Outputs
    Target Audience:The target audience is academic, amateur and the public at large. Global warming has become the dominant scientific paradigm in vegetation science, shifting the focus from the understanding of steady-state ecological functioning and dynamics to future change of ecosystems. But one is dependent on the other. An important question is whether it is prudent to bypass empirical research toward futuristic modeling experiments that are also based on another layer of climate change modeling. Where does empiricism begin and end? These predictive efforts--time-to-space substitution--risk similar hazards of time-to-space substitution which plague retrospective studies similarly challenged by historical data. The specific hypothesis at hand is whether future climate change will drive plant species and assemblages both poleward and to higher elevations. Without real time verification, vegetation forecast models are an untested exercise; proper science requires empirical testing of the global warming hypothesis. Mapping the complexity of nature at high resolution has recently been made possible using Google Earth which permits rapid efficient inventory both spatially and temporally. Google Earth has also become the library of aerial imagery for the world, and this database will grow rapidly in the time frame of 21st century global climate model simulations. The maps in this atlas should be viewed as a single time slice in an evolutionary process of building upon previous work, through continual refinement with improvement of imagery quality and for long-term vegetation change. Vegetation science should follow the map-making tradition of bedrock substrate inventory in the geological sciences since William Smith in the early 19th century. The maps in this research can be refined or redefined for a wide range of research hypotheses because this analysis is shown with the original vegetation on Google Earth imagery. Revisions can be efficiently undertaken because aerial imagery record is archived as a library on the Google Earth platform, and new data can be entered electronically. The maps represent the first comprehensive revision since the Vegetation Type Map Survey (VTM) of 1929-34 in California and the first comprehensive baseline for the Baja California peninsula. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The methods and approaches of this research can be applied anywhere because of Google Earth's comprehensive global coverage of imagery. The analysis in this atlas led to an evolutionary development of methods that would permit broad scale inventory of tree species anywhere. While this effort did not directly lead to training and professional development at UCR, there is great potential to expand professional training elsewhere. Every region in the world has ecologists with skilled local background in imagery interpretation. How have the results been disseminated to communities of interest?It is the intent to disseminate the results of this research in a single electronic book to test and refine maps using a "community on-line atlas" using the Google earth platform. In an electronic format that georeferences this vegetation polygon analysis with the original vegetation data on Google Earth, maps can be taken into the field or in the laboratory using mobile electronic devices and computers. Maps can be reconstructed and redefined at high resolution for individual purposes, and with the history function one can evaluate the fate of individuals and plant assemblages over time. What do you plan to do during the next reporting period to accomplish the goals?Based on the informal review of my book manuscript, I will map tree distributions in the Sierra Nevada, northern Mojave Desert, and the San Joaquin Valley. One consequence will be the addition of nine new species found exclusively in these areas, giving a total of 85 tree species for the entire atlas. The book manuscript will be revised to accommodate new map data.

    Impacts
    What was accomplished under these goals? With the posting of new imagery on Google Earth, tree species mapshave been expanded to include the three Californias along the Pacific coast: California (north to the latitude of San Francisco), as well as Baja California and Baja California Sur in Mexico. The enlarged mapping area of this survey permits the completion of California tree species southward along the Baja California peninsula instead of the US-Mexico political boundary. The area was also expanded because free-burning (no fire management) in Mexico results in different biomass burning outcomes compared to that under fire suppression in California. At this writing, maps of 76 species were completed as well as the preparation of first draft of a book entitled Tree Atlas of the Californias. The manuscript is organized accordingly: Google Earth and vegetation science, Geology and climate, Cenozoic climate and paleo vegetation history, Botanical exploration, Google Earth methods, Atlas of tree species, Species changes since the last glacial maximum, and Conclusions. For each species, new maps with unprecedented resolution are presented with discussion of the late Cretaceous and Cenozoic fossil record, patterns of disturbance, especially from fires and floods, and changes in distribution since the last ice age maximum 20,000 years ago. The final maps are generally consistent with the VTM survey 80 years ago. However, this analysis also resulted in 25 range expansions of well-known tree species that are unknown to science, largely because Google Earth imagery provides high resolution data in inaccessible terrain. The manuscript was read informally by a faculty at the University of California, Riverside. The primary criticism was to expand this analysis to include the California Sierra Nevada, northern Mojave Desert and the San Joaquin Valley (northward to lat. 38.05 °N), an area that had low quality imagery until very recently. The completion of this atlas will require the addition of map data in these areas.

    Publications

    • Type: Book Chapters Status: Published Year Published: 2016 Citation: Minnich, R.A., B.R. Goforth, and T.D. Paine. Follow the water: Extreme drought and the conifer forest pandemic of 2002-2003 along the California borderland. In T.D. Paine, F. Lieutier (eds.) Insects and diseases of Mediterranean Forest Systems, DOI 10.1007/978-3-319-24744-1_29


    Progress 10/01/16 to 09/30/17

    Outputs
    Target Audience:The target audience is academic, amateur and the public at large. Global warming has become the dominant scientific paradigm in vegetation science, shifting the focus from the understanding of steady-state ecological functioning and dynamics to future change of ecosystems. But one is dependent on the other. An important question is whether it is prudent to bypass empirical research toward futuristic modeling experiments that are also based on another layer of climate change modeling. Where does empiricism begin and end? These predictive efforts--time-to-space substitution--risk similar hazards of time-to-space substitution which plague retrospective studies similarly challenged by historical data. The specific hypothesis at hand is whether future climate change will drive plant species and assemblages both poleward and to higher elevations. Without real time verification, vegetation forecast models are an untested exercise; proper science requires empirical testing of the global warming hypothesis. Mapping the complexity of nature at high resolution has recently been made possible using Google Earth which permits rapid efficient inventory both spatially and temporally. Google Earth has also become the library of aerial imagery for the world, and this database will grow rapidly in the time frame of 21st century global climate model simulations.The maps in this atlas should be viewed as a single time slice in an evolutionary process of building upon previous work, through continual refinement with improvement of imagery quality and for long-term vegetation change. Vegetation science should follow the map-making tradition of bedrock substrate inventory in the geological sciences since William Smith in the early 19th century. The maps in this research can be refined or redefined for a wide range of research hypotheses because this analysis is shown with the original vegetation on Google Earth imagery. Revisions can be efficiently undertaken because aerial imagery record is archived as a library on the Google Earth platform, and new data can be entered electronically. The maps are the first comprehensive baseline since the Vegetation Type Map Survey (VTM) of 1929-34. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The methods and approaches of this research can be applied anywhere because of Google Earth's comprehensive global coverage of imagery. The analysis in this atlas led to an evolutionary development of methods that would permit broad scale inventory of tree species anywhere. While this effort did not directly lead to training and professional development at UCR, there is great potential to expand professional training elsewhere. Every region in the world has ecologists with skilled local background in imagery interpretation. How have the results been disseminated to communities of interest?It is the intent to disseminate the results of this research in a single electronic book to test and refine maps using a "community on-line atlas" using the Google earth platform. In an electronic format that georeferences this vegetation polygon analysis with the original vegetation data on Google Earth, maps can be taken into the field or in the laboratory using mobile electronic devices and computers. Maps can be reconstructed and redefined at high resolution for individual purposes, and with the history function one can evaluate the fate of individuals and plant assemblages over time. What do you plan to do during the next reporting period to accomplish the goals?Based on the informal review of my book manuscript, I will map tree distributions in the Sierra Nevada, northern Mojave Desert, and the San Joaquin Valley. One consequence will be the addition of nine new species found exclusively in these areas, giving a total of 85 tree species for the entire atlas. The book manuscript will be revised to accommodate new map data.

    Impacts
    What was accomplished under these goals? With the posting of new imagery on Google Earth, tree species mapshave been expanded to include the three Californias along the Pacific coast: California (north to the latitude of San Francisco), and Baja California and Baja California Sur in Mexico. The enlarged mapping area of this survey permits the completion of California tree species southward along the Baja California peninsula instead of the US-Mexico political boundary. The area was also expanded because free-burning (no fire management) in Mexico results in different biomass burning outcomes compared to that under fire suppression in California. At this writing, maps of 76 species were completed as well as the preparation of first draft of a book entitled Tree Atlas of the Californias. The manuscript is organized accordingly: Google Earth and vegetation science, Geology and climate, Cenozoic climate and paleo vegetation history, Botanical exploration, Google Earth methods, Atlas of tree species, Species changes since the last glacial maximum, and Conclusions. For each species, new maps with unprecedented resolution are presented with discussion of the late Cretaceous and Cenozoic fossil record, patterns of disturbance, especially from fires and floods, and changes in distribution since the last ice age maximum 20,000 years ago. The final maps are generally consistent with the VTM survey 80 years ago. However, this analysis also resulted in 25 range expansions of well-known tree species that are unknown to science, largely because Google Earth imagery provides high resolution data in inaccessible terrain. The manuscript was read informally by a faculty at the University of California, Riverside. The primary criticism was to expand this analysis to include the California Sierra Nevada, northern Mojave Desert and the San Joaquin Valley (northward to lat. 38.05 °N), an area that had low quality imagery until very recently. The completion of this atlas will require the addition of map data in these areas.

    Publications


      Progress 10/01/15 to 09/30/16

      Outputs
      Target Audience:Since the final product will be a series of maps in an atlas and the project establishes the use of Google Earth as a basis of mapping analysis, the intended target audience will be broad and will include academic professionals in ecology, biogeography, taxonomy plant pathology and entomology, as well as in remote sensing and geographic information systems. The atlas is also targeted for those involved of land conservation, management and planning both in the public and private sectors. Maps will also attract amateur interest in the biological sciences. Changes/Problems:As previously stated, this project has expanded from southern California to cover central California west of the central Valley and the Baja California peninsula. The enlarged area of range has increased the insight of species-environmental relationships in California and the adjacent Baja California peninsula. What opportunities for training and professional development has the project provided?I have self-trained myself and created vegetation maps for most of my career. The Google Earth platform has allowed me to generate vegetation maps at much greater speed and accuracy than under separated remote sensing and geographic information systems technologies. The union of remote sensing and geographic information systems is the crowning achievement of Google Earth. How have the results been disseminated to communities of interest?To date I have given seminars on Google Earth vegetation analysis on campus at the University of California Riverside. Before this project I had published vegetation maps using Google Earth as part of research studies on palm oases, and chaparral in Baja California. What do you plan to do during the next reporting period to accomplish the goals?For the coming reporting period, my primary effort will focus on the preparation and completion of a book manuscript for publication. To date the writing of the first draft is nearly complete. Species range maps need to be prepared to publication standards. My goal is to have this atlas published in an electronic format so that maps can be shown using the Google Earth platform. This will allow the reader to associate polygon data with the raw data on imagery, just as I saw it in my survey. Not only will the reader be able to view species ranges, but also recognize them from their morphological properties seen on imagery (an aerial photograph trainer). Species properties are also described in reference tables. This effort can be a model for vegetation mapping in other regions globally.

      Impacts
      What was accomplished under these goals? Maps of 77 tree species have been completed or at near completion. A few species were recently revised with the posting of new high quality imagery on Google Earth. Because many California tree species ranges extend past the US-Mexican boundary, mapping was also extended to include California species ranges to their ultimate southern limits in the Baja California peninsula. This led to the mapping of members of Mexican pine-oak woodlands and palm oases species found in the southern and central peninsula of Baja California using Google Earth imagery. The product comprises unprecedented unparalleled high resolution maps. Compared with maps based on ground-based botanical collections and vegetation surveys, the vantage and "access" into remote regions provided by aerial photography and space imagery of Google Earth in California and Baja California has resulted in the discovery of new populations and range expansions of many species. This survey has resulted in the documentation of 19 range expansions at localities far away of known populations, 9 in California and 10 in Baja California. The survey has also produced the first maps interpreted from aerial photography for 8 species in California, entirely from areas not mapped in the 1929-34 VTM Survey on the southern California Channel Islands and the northeastern Mojave Desert. The distributions of two similarly looking species mismapped by the VTM survey (Ponderosa pine, Jeffrey pine) were substantially revised for the San Gabriel, San Bernardino and San Jacinto Mountains of southern California. This survey also produced the first range maps based on aerial imagery of 31 tree species in the Baja California peninsula. Some new ranges will influence future study of several species. For example, the range of Arizona cypress, thought to be a rare species, was found to be extensive in the mountains of northern Baja California. This finding has implications on the taxonomy of the genus. Taken a a whole, this survey brings a new vegetation baseline and insight on several issues including (1) ecological segregation of deciduous and evergreen trees in terms of soils and bedrock. (2) ecological segregation of California oak woodland from Mexican oak woodland species. (3) importance of resistant bedrock slopes and bedrock runoff that maintain local populations of "moist" tree species in dry climate. (4) differences in the spatial and temporal dynamics of tree species in response to fire regimes. (5) long-term persistence of species ranges since initial colonization of California in the 18th century in spite of climate change and grazing. (6) importance of the climate transition on species turnover, from winter rain in California and northern Baja California to summer rain in central and southern Baja California. (7) effect of fire suppression on species changes in California mixed conifer forests, but not in Baja California under free-burning. (8) influence of summer cold fog and fog drip on several trees restricted to the California coast line. (9) destructive impact of introduced pathogens on several tree species. The analysis here should be viewed in the tradition of the geological sciences for which maps of surface bedrock have been revised and refined over two centuries. It is expected that the maps produced in this project will be revised due to error and omissions, as well as long-term vegetation change. Moreover, it is also expected that variations in map analysis by other workers depending on their research goals. The time slice of species ranges in this atlas can be compared with past and future efforts. The high resolution of species maps will provide an empirical database for testing vegetation change against global climate change.

      Publications


        Progress 06/23/15 to 09/30/15

        Outputs
        Target Audience: Nothing Reported Changes/Problems:The decision to enlarge the scale of mapping tree species to include the central California and the Baja California peninsula represents the benefits of using Google Earth in vegetation mapping. The methodologies developed in this survey can be applied to vegetation analysis everywhere. What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?In the next period, I planned to produce high resolution maps of 75 species in the Californias (California, Baja California, and Baja California Sur.

        Impacts
        What was accomplished under these goals? This report covers the beginning phases of this project. Protocols are being developed using Google Earth to create high resolution tree species maps. In the process it was discovered that Google Earth imagery and programming allows for efficient survey of each species. As a consequence, the original southern California region planned for analysis was enlarged to include the coast ranges and valleys of central California west of the San Joaquin Valley, as well as the Mojave and Colorado Deserts of southeastern California. With heretofore ready access of aerial imagery of Baja California and Baja California Sur in Mexico, the ranges to these species will be completed into the Baja California peninsula. Some palm and oak species endemic to Baja California will also be mapped. The methodology of aerial photograph interpretation using manual raster scanning was refined upon completion of several species ranges including Juniperus grandis, Pinus ponderosa, and Pinus muricata. High resolution maps for all species will exceed the quality and information of all published maps of the 75 species planned in this effort. Completed maps of these 3 species already reveal discovery of range extensions and error in previous work.

        Publications